基于支持向量机的复杂岩性测井识别方法  

A new identification method for complex lithology with support vector machine

作　　者：宋延杰[1] 张剑风[1] 闫伟林[2] 何英伟[3] 王德平[4] 

机构地区：[1]大庆石油学院地球科学学院,黑龙江大庆163318 [2]大庆油田有限责任公司勘探开发研究院,黑龙江大庆163712 [3]大庆油田有限责任公司第五采油厂,黑龙江大庆163513 [4]大庆油田有限责任公司测试分公司,黑龙江大庆163517

出　　处：《大庆石油学院学报》2007年第5期18-20,46,共4页Journal of Daqing Petroleum Institute

摘　　要：在复杂岩性自动识别中,聚类分析方法只有在样本趋于无穷大时,才能从理论上保证结果的精度,神经网络容易陷入局部最小,而使用范围受到限制.针对复杂岩性识别中所面临的问题,提出了在解决模式识别小样本、非线性及高维问题中具有独特优势的支持向量机方法来识别岩性.通过对巴彦塔拉油田部分层段的样本进行学习和预测,并与实际取心资料进行对比,符合率平均值为96%,表明支持向量机在岩性识别中可获得良好的应用效果.In the research of identification method for complex lithology, cluster analysis obtains high accuracy only when the number of the samples is infinite in theory. The application of neural network is limited because of its algorithm flaw. Because Support Vector Machine (SVM) can obtain high accuracy in less samples and nonlinear multi-dimension pattern recognition problem, this method is used in order to resolve the problem of identification of complex lithology in this paper. Core data is used to establish SVM model, and this model is used to predict formation lithology in Bayantala oil filed. The correctness of lithology identification is 96%. The result shows that SVM can perfectly resolve identification problem of complex lithology.

关 键 词：岩性识别 测井解释 支持向量机(SVM) 模式识别 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]